发表机构
Chungbuk National University; University of Innsbruck（忠北国立大学; 因斯布鲁克大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对斜向查询中潜在属性难以表面匹配的问题，提出OBLIQ-IR稠密检索器，通过合成查询与kNN图蒸馏训练，显著提升各任务NDCG@10。
AI 中文摘要
斜向检索，如OBLIQ-Bench所示，要求检索器找到相关性由潜在属性（隐含立场、类比推理技巧、作者指纹或模糊的舌尖回忆）决定的文档，而这些属性在文档中几乎没有或完全没有表面表达。最先进的稠密编码器和围绕前沿语言模型构建的智能体搜索管道在这些任务上表现出巨大的第一阶段瓶颈，而同样的语言模型在展示候选文档时能够可靠地验证相关性。我们通过OBLIQ-IR解决这一问题，这是一种单向量稠密检索器，其训练混合了按机制生成的合成查询与一种新的跨模型监督形式：来自冻结作者编码器的k近邻图蒸馏，将风格与主题的归纳偏差转移到学生模型中。微调后的3B检索器在Writing-Style上达到0.211 NDCG@10，在Math上达到0.171，在Twitter上达到0.177，在Congress上达到0.281，在每项报告任务上比GPT-5.2多跳智能体提高0.010至0.150 NDCG@10，比Gemini-2-Embedding提高0.027至0.222 NDCG@10。代码、数据和检查点可从此https URL获取。
英文摘要
Oblique retrieval, as exemplified by OBLIQ-Bench, asks a retriever to find documents whose relevance is determined by a latent attribute (an implicit stance, an analogous reasoning technique, an authorial fingerprint, or a vague tip-of-the-tongue recollection) that has little or no surface expression in the document. State-of-the-art dense encoders and agentic search pipelines built around frontier language models exhibit a large first-stage bottleneck on these tasks, while the same language models reliably verify relevance when shown candidates. We address this with OBLIQ-IR, a single-vector dense retriever whose training mixture combines per-mechanism synthetic queries with a new form of cross-model supervision: kNN-graph distillation from a frozen authorship encoder, which transfers a style-versus-topic inductive bias into the student. A 3B retriever fine-tuned reaches 0.211 NDCG@10 on Writing-Style, 0.171 on Math, 0.177 on Twitter, and 0.281 on Congress, improving over the GPT-5.2 Multi-Hop Agent by \xr{0.010 to 0.150} NDCG@10 and over Gemini-2-Embedding by 0.027 to 0.222 NDCG@10 on every reported task. The code, data and checkpoints are available https://github.com/DataScienceUIBK/obliq-ir
CommentsAccepted at MAIN EMNLP 2026
